If you have read my previous article in this series, then you will be familiar with the SMART acronym as applied to the research question. I apply it slightly differently when considering the research goals.
A specific research objective must not only be precisely defined, but it must deliver a specific component of the final PhD.
A measurable objective is one that can be demonstrated to have delivered its specified component
An achievable objective is a component of an achievable whole: it only takes one objective to be unachievable for the whole PhD to fail.
The relevance criteria of an objective defines the scope of that component of the study. All activity must be relevant to at least one objective, or it should be abandoned. One of the most significant risks to a PhD is following a red herring into a cul de sac. During the study you need to use your objectives to ask yourself "Is this a relevant activity?" If not, Stop. At once!
Objectives should be mapped onto specific phases of activity all of which should be time limited with a start and stop date. Some of these will run in parallel, some may even run throughout the project. For example, literature review is an activity which will have an intense initial phase, followed by an ongoing monitoring phase for new literature, and then finally a check before submission to make sure you have not missed something really important.